|
Activity Number:
|
68
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Sunday, August 2, 2009 : 4:00 PM to 5:50 PM
|
|
Sponsor:
|
Section on Nonparametric Statistics
|
| Abstract - #303337 |
|
Title:
|
Estimation of Shift Based on Smoothed Kolmogorov-Smirnov Statistics
|
|
Author(s):
|
Feridun Tasdan*+
|
|
Companies:
|
Western Illinois University
|
|
Address:
|
Department of Mathematics, Macomb, IL, 61455,
|
|
Keywords:
|
Shift Parameter ; Two-Sample Problem ; Kolmogorov-Smirnov ; Smoothing
|
|
Abstract:
|
A new method of estimating a shift parameter is the main topic of this study. The new method smoothes the empirical distribution functions used in the traditional Kolmogorov-Smirnov method to find a smoothed test statistics which can be minimized with respect to an arbitrary shift variable in order to estimate the true shift parameter. The asymptotic properties of the new estimator are investigated such as asymptotic distribution, and asymptotic level confidence interval. A bootstrap simulation study is performed to compare the proposed method with existing shift parameter estimators such as Hodges-Lehmann or Least Squares. The result of the simulation study shows that the new estimator can compete with the existing estimators under some conditions. Further improvements to the proposed method are also discussed.
|
